This is my personal blog. I work at Microsoft as a Consultant. You can find some of my open source projects on Github. I like to share my knowledge about AI, Azure, PowerShell and DevOps
Introduction I always like to see which tools people are using in their day to day activities. In the past I watched for the blog posts from Scott Hanselman and the tools he is using. He has not published any new list of tools lately, but one of the latest tools he showed during our internal Microsoft MCAPS Tech Connect event was Handy. It is a free and open source app for speech to text, whic...
Introduction I have been running two experiments side by side for a while now. One is a personal knowledge wiki that I want both me and my agents to read. The other is Scout, Microsoft’s enterprise take on the OpenClaw-style always-on agent, where a small crew of named AI teammates helps me get work done. This week I finally wired the two together, and it turned out to be a lot more useful tha...
Introduction I was at AgentCon Utrecht yesterday. Eduard van Valkenburg a Senior Software Engineer at Microsoft - CoreAI and Semantic Kernel & Agent Framework team demoed the Microsoft Agent Framework v1.0, and the Harness Agent pattern was the thing that stuck with me most on the drive home. I’ve been building agentic solutions for a while now, from CrewAI investment analysis workflows t...
TL;DR I used the LLM Council technique to have all 17 GitHub Copilot CLI models anonymously evaluate themselves and each other. Key findings: Opus models = deep reasoning, architecture decisions, “the expensive senior engineer” Sonnet models = balanced capability/speed, good default for most coding tasks Haiku / Mini models = fast execution for simple, well-defined tasks Codex models...
Introduction I recently bought the book Designing Multi-Agent Systems: Principles, Patterns, and Implementation for AI Agents by Victor Dibia. In chapter 10 “Evaluating Multi-Agent Systems” he discusses how to build evaluation frameworks with LLM-as-judge and metrics. This was a new topic for me and wanted to learn more about what Agent Evaluators are and why they matter. This is why I wrot...